Week 9 - Decision Theory

Question 1

How do you track decisions?

Answer 1

Decision log
- Who
-When
- Why

Question 2

Routine Decision

Answer 2

• small impact
• Reversible
• short term
• data available
• standard decision process

Question 3

Make-or-break decision

Answer 3

High impact
irreversible
long term
safety or product liability is critical
no well defined decision process

Question 4

Structured decision processes

Answer 4

• Qualitative
• Quantitative

Question 5

How is risk described?

Answer 5

A - Future Event
C - Activity consequence
C* - Prediction of C
U - Uncertainty about what value C will take
P - probability given K
K - background information

Question 6

Risk analysis methods

Answer 6

• Simplified (Qualitative )
• Standard (Quantitative or qualitative)
• Model-based (primarily quantitative)

Question 7

simplified risk analysis

Answer 7

informal procedure using brainstorming and discussion

Question 8

Standard Risk analysis

Answer 8

Formalized procedure where recognized risk analysis methods are used.
Risk matrices are often used.

Question 9

Model-based risk analysis

Answer 9

Makes use of techniques such as event tree and fault tree analysis.

Question 10

Fault Tree Analysis (FTA)

Answer 10

System analysis techniques used to determine the root causes and probability of occurrence of a specified undesired event.

Question 11

OR gate probability

Answer 11

P(A U B) = P(A) + P(B) - P(A)P(B)

Question 12

OR gate probability for mutually exclusive events

Answer 12

P(A U B) = P(A) + P(B)

Question 13

OR gate with 3 inputs Probability

Answer 13

P(A+B+C) = P(A)+P(B)+P(C) - P(A)P(B)-P(A)P(C)-P(B)*P(C) + P(A)P(B)P(C)

Question 14

AND Gate probability

Answer 14

P(A ∩ B) = P(A)P(B)

Question 15

Cut sets

Answer 15

Sets of Events that together cause the top undesired event to occur.

Question 16

What does a low-order cut set indicate?

Answer 16

High safety vulnerability

Question 17

Reactive FTA

Answer 17

Used after an accident as an investigation method

Question 18

Proactive FTA

Answer 18

Performed during system development to predict and prevent future problems.

Question 19

Event Tree analysis (ETA)

Answer 19

Inductive Procedure that shows all possible outcomes resulting from an accidental initiating event.

Question 20

How is ETA useful?

Answer 20

• identifies potential accident scenarios
• scenario frequencies
• scenario consequences

Question 21

ETA Structure

Answer 21

initiating event
pivotal events
outcomes
probability
consequences

Question 22

For ETA:
Multiply along…
Sum across…
set of branches probability must sum to…

Answer 22

multiply along horizontal branches
sum across vertical branches
must sum to 1

Question 23

Uses of Markov Chains

Answer 23

• Predicting traffic flows
• communications networks load
• genetic issues
• currency exchange rates
• population dynamics

Question 24

stationary matrix

Answer 24

Matrix converges to steady state
Called a Regular Markov Chain

Question 25

How to tell if a matrix is regular?

Answer 25

All entries become positive

Question 26

Absorbing state

Answer 26

If the state is impossible to leave once it is entered

Question 27

Requirements for an absorbing Markov Chain

Answer 27

Minimum of 1 absorbing state It is possible to go from each non-absorbing state to at-least one absorbing state

Question 28

What makes a decision sensitive

Answer 28

- After a test the decision is different for different test results. - The decision changes after more tests

Question 29

Reasons to not do a test for a decision

Answer 29

- avoid costs if result is unlikely to affect decision - If multiple tests are available a different one may be chosen.

Question 30

Pre-posterior

Answer 30

doing a test, updating the probability distribution and deciding what to do next.

Question 31

Posterior (Bayes)

Answer 31

Involves calculating probabilities and deciding on action given an experiment has already been done. (result is known)

Question 32

Types of uncertainty

Answer 32

aleatory - random Epistemic - Lack of Knowledge

Question 33

Type 1 Error

Answer 33

Rejecting a True Hypothesis (False Positive)

Question 34

Type 2 Error

Answer 34

Accepting a False Hypothesis (False Negative)

Question 35

Utility

Answer 35

Defined as the state of being satisfied, useful, beneficial etc... Ranks happiness

Question 36

Marginal Utility

Answer 36

Additional utility gained from the consumption of one additional unit.

Question 37

Total utility

Answer 37

Total amount of satisfaction that a person/decision maker can receive from the consumption of all units of a good or service.

Question 38

Law of Diminishing Marginal Utility

Answer 38

As you consume more you get diminishing utility but you still consume the same number of resources.

Question 39

Risk Neutral

Answer 39

bases all decisions on expected utility

Question 40

Risk Averse

Answer 40

Will not be comfortable with a large probability of big loss

Question 41

Not Risk averse

Answer 41

Will be comfortable with a small probability of a big win

Question 42

Latin Hypercube sampling

Answer 42

Decide how many sample points to use. Each sample point the row and column it was taken from must be recorded

Question 43

Orthogonal Sampling

Answer 43

Sample space is divided into equal subspaces. All points are chosen simultaneously ensuring that each subspace is sampled with the same density